OPTIMIZING TWO-STORAGE INVENTORY MANAGEMENT FOR LOG-GAMMADECAYING GOODS WITH QUADRATIC DEMAND: A GENETIC ALGORITHMAPPROACH

Abstract

This paper presents a sophisticated model for the optimal management of inventories comprising
deteriorating items stored in two distinct warehouses. The model encompasses a nuanced treatment
of shortages, utilizing a genetic algorithm to implement partial backlogging, where demand is
contingent upon both selling price and time. In instances where the ordered quantity exceeds the
primary warehouse's capacity, any surplus stock is strategically allocated to a rented warehouse.
To minimize storage costs, the genetic algorithm prioritizes the release of items from the rented
warehouse. Consequently, the stock in the rented warehouse gradually depletes to zero over
intervals due to demand and deterioration, while items in the owned warehouse decrease solely
due to deterioration. After a predetermined timeframe, the inventory level in the owned warehouse
reaches zero, initiating shortages.
The model assumes that both the rate of backlogging and demand follow generalized exponential
decreasing functions with respect to selling price (p) and time (t). Numerical examples are
employed to illustrate the application of the genetic algorithm-based model, showcasing its
efficacy under diverse scenarios. Additionally, sensitivity analysis is conducted to scrutinize the
model's behavior under various parameter variations.
Keywords: Inventory management, deteriorating items, two warehouses, Shortages, Partial
backlogging, Selling price, Time, Genetic algorithm